Data wydarzenia:

What’s the Meaning of Superhuman Performance in Natural Language Understanding?

Data: wtorek, 12.12.2023, godz. 11:00-12:00

Prelegent: Roberto Navigli, Simone Tedeschi (Sapienza University of Rome)

Streszczenie: Lately, Natural Language Processing (NLP) has impressed the general public with the introduction of generative AI systems based on pre-trained Large Language Models (LLM). The astonishing results of these systems have led researchers to repeatedly flaunt the achievement of "superhuman performance". Beyond the shared feelings and extraordinary achievements of LLMs in a large number of tasks, in this talk I will analyze the current landscape carefully and critically, highlighting both the low-hanging fruits and the exaggerations of enthusiasts and, consequently, the weaknesses and areas for improvement.

Miejsce: B1-7/8 oraz online